This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Lecture 10 2/08/08 Chi-Square Test We flip a coin with some tape on one side 100 times and observe the result. We observe 60 heads flips and 40 tail flips. We expected 50 head flips and 50 tail flips. Did the tape make a difference? Particularly suited to dealing with ratio data. (ex: 9:3:3:1 or 3:1) Chi Square Groups Obs Exp O-E (O-E)^2 (O- E)^2/E Heads 60 50 10 100 2 Tails 40 50-10 100 2 Sum of last column = 4 Square will get rid of the negative numbers. 1. Get Data 2. Do math 3. Number that comes out is called the statistic. Statistic is not a procedure. The test is the procedure. 4. The chi square test statistic is the number that comes out. 5. The next step is to take the statistic and look up the probability due to chance in a table. 6. Df stands for degrees of freedom. Calculate degrees of freedom. Number of groups minus one. 7. 0.05 > p > 0.025 8. Table wraps completely around. If greater than the last number or less than the smallest number, then the probability is less than 0.005 and 0.995 respectively. The probability that we look up in the Chi square table (p.99) has the following definition (ww): It is the probability that we could get a deviation between the expected and observed results as...
View Full Document
- Spring '08